MLSP-47.2
WHEN DOES BACKDOOR ATTACK SUCCEED IN IMAGE RECONSTRUCTION? A STUDY OF HEURISTICS VS. BI-LEVEL SOLUTION
Vardaan Taneja, Indian Institute of Technology, Delhi, India; Pin-Yu Chen, IBM Research, United States of America; Yuguang Yao, Sijia Liu, Michigan State University, United States of America
Session:
Deep Learning VII
Track:
Machine Learning for Signal Processing
Location:
Gather Area F
Presentation Time:
Fri, 13 May, 20:00 - 20:45 China Time (UTC +8)
Fri, 13 May, 12:00 - 12:45 UTC
Fri, 13 May, 12:00 - 12:45 UTC
Session Chair:
Yanmin Qian, Shanghai Jiaotong University
Session MLSP-47
MLSP-47.1: PREDICTING THE GENERALIZATION GAP IN DEEP MODELS USING ANCHORING
Vivek Narayanaswamy, Andreas Spanias, Arizona State University, United States of America; Rushil Anirudh, Irene Kim, Yamen Mubarka, Jayaraman J Thiagarajan, Lawrence Livermore National Laboratory, United States of America
MLSP-47.2: WHEN DOES BACKDOOR ATTACK SUCCEED IN IMAGE RECONSTRUCTION? A STUDY OF HEURISTICS VS. BI-LEVEL SOLUTION
Vardaan Taneja, Indian Institute of Technology, Delhi, India; Pin-Yu Chen, IBM Research, United States of America; Yuguang Yao, Sijia Liu, Michigan State University, United States of America
MLSP-47.3: MIXED KNOWLEDGE RELATION TRANSFORMER FOR IMAGE CAPTIONING
Tianyu Chen, Zhixin Li, Jiahui Wei, Tiantao Xian, Guangxi Normal University, China
MLSP-47.4: BALANCED STRIPE-WISE PRUNING IN THE FILTER
Zheng Huo, Chong Wang, Weiwei Chen, Yuqi Li, Ningbo University, China; Jun Wang, China University of Mining and Technology, China; Jiafei Wu, SenseTime Research, China
MLSP-47.5: GAN-BASED JOINT ACTIVITY DETECTION AND CHANNEL ESTIMATION FOR GRANT-FREE RANDOM ACCESS
Shuang Liang, Yinan Zou, Shanghaitech University, Shanghai Institute of Microsystem and Information Technology and University of Chinese Academy of Sciences, China; Yong Zhou, Shanghaitech University, China